1. Field of the Invention
The present invention relates to an image processing apparatus and an image processing method, which process images acquired by a capsule endoscope that is introduced into a subject and captures images of inside of the subject.
2. Description of the Related Art
In recent years, examinations using capsule endoscopes that are introduced into subjects such as patients and capture images of inside of the subjects are known in the field of endoscopes. A capsule endoscope is an apparatus that has built-in imaging function, wireless communication function, and the like, which are in a casing of a capsule shape that is formed into a size introducible into a digestive tract of a subject, and sequentially, the capsule endoscope wirelessly transmits image data generated by capturing images of inside of the subject to outside of the subject. The series of image data wirelessly transmitted from the capsule endoscope are once accumulated in a receiving device provided outside of the subject, are transferred to (downloaded by) an image processing apparatus such as a work station from the receiving device, and are subjected to various image processing in the image processing apparatus. Thereby, a series of images, in which an organ or the like in the subject is photographed, are generated.
These images are displayed on a screen in a format of a pseudo moving image, a list of still images or the like. A medical worker (medical doctor) performs diagnosis on the subject by observing the images displayed on the screen and selecting those with an abnormality.
One examination using the capsule endoscope takes about eight hours and the images acquired during that period of time amount to about 60,000 images. Therefore, to observe all of these images takes a very long time and requires an ability to concentrate. Accordingly, in the image processing apparatus, in order to aid the work of observation, an abnormal part extraction function, which is for automatically extracting images considered to have a medical abnormality, for example, like images having high red components, is available. This abnormal part extraction function is realized generally by collecting images determined by medical workers to have a medical abnormality and incorporating software, which has an algorithm of extracting images with feature data similar to those images, into hardware such as a work station.
As a technique related to display of images acquired by examinations, for example, in Japanese Patent Application Laid-open No. 2005-288043, a medical image diagnosis apparatus is disclosed, which displays, together on a same screen, diagnosis images being captured in real time at a medical diagnosis apparatus, such as a CT apparatus, an MR apparatus, an ultrasonic wave diagnosis apparatus, or the like and diagnosis images that have been captured in the past.